{"id":"W4414123053","doi":"10.36227/techrxiv.175760290.00742839/v1","title":"Domain Knowledge is Power: Leveraging Physiological Priors for Self-Supervised Representation Learning in Electrocardiography","year":2025,"lang":"en","type":"preprint","venue":"","topic":"ECG Monitoring and Analysis","field":"Medicine","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Southeastern Ontario Academic Medical Organization; Canadian Institutes of Health Research; Social Sciences and Humanities Research Council of Canada; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Prior probability; Domain (mathematical analysis); Representation (politics); Domain knowledge; Electrocardiography; Feature learning","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004943642,0.0003336976,0.0009184264,0.0008030398,0.000117857,0.00004715744,0.0001706655,0.0003909871,0.000046286],"category_scores_gemma":[0.000101984,0.0002921529,0.0009506677,0.0007869596,0.00002686643,0.0000353625,0.0002365122,0.001026413,0.000008225295],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001754564,"about_ca_system_score_gemma":0.0002076412,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001017831,"about_ca_topic_score_gemma":0.000003896645,"domain_scores_codex":[0.9977885,0.0001921475,0.0004955154,0.0009012433,0.0002141824,0.0004083841],"domain_scores_gemma":[0.9989327,0.0001968932,0.0001228317,0.0004572872,0.0001958255,0.00009446277],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0010771,0.001914864,0.877881,0.004521671,0.007064906,0.00007179142,0.02049376,0.003397667,0.02549891,0.0005356348,0.01517654,0.04236617],"study_design_scores_gemma":[0.03221373,0.003742883,0.5049661,0.01259501,0.009332893,0.00004246184,0.02664188,0.2499145,0.05058249,0.02291849,0.08018706,0.006862525],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9580469,0.001359429,0.02202951,0.001386313,0.0003988015,0.001327654,0.000006954494,0.0006303989,0.01481403],"genre_scores_gemma":[0.9649183,0.0006600918,0.02991424,0.0002229265,0.0003713029,0.00029779,0.0001695122,0.00002916438,0.003416743],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3729149,"threshold_uncertainty_score":0.999953,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0296075816117653,"score_gpt":0.337165800434429,"score_spread":0.3075582188226637,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}